Videos corresponding to this chapter are available on DVD and online.
RATIONALE AND BASIC SCIENCE PERTINENT TO THE PROCEDURE
∗ I would like to dedicate this chapter to my terrific children, David, Kimberly, and Carolyn, and to my loving parents, Berish and Rena.Although pain serves a useful function in preventing bodily injury (witness those with congenital insensitivity to pain, diabetics, and so on), chronic pain can be debilitating—both physically and mentally. With respect to the wrist joint, the chief indication for most nonacute surgical procedures is pain. Pain relief is usually afforded by restoring or correcting anatomy or by salvage operations that distort normal anatomy (fusions, resections) but have the advantage of reducing or eliminating pain.
An alternative or adjunct to traditional wrist salvage procedures such as fusions and resections is wrist denervation. Ultimately, pain is mediated by nerve endings that innervate the wrist joint and periarticular structures. Sectioning these articular nerve branches theoretically reduces pain without causing motor or sensory deficits or affecting motion. Good to excellent long-term results have been reported with wrist denervation procedures. It is interesting that Charcot joints have never been reported following wrist denervation, probably because of the inability to produce complete wrist denervation.
John Hilton (1804–1878) was a British anatomist and surgeon who noted that nerves crossing a joint give off branches to innervate that joint. “Hilton’s law,” bearing his name, embodies that concept. Cutting these articular branches as a means of reducing pain was apparently first proposed by Camitz in 1933, who suggested denervating hip joints for chronic osteoarthritis. Tavernier in 1949 reported excellent results in 75% of patients undergoing hip joint denervation. In 1954, Marcacci proposed denervating knee joints, and Nyakas and Kiss at the same time expanded the indications to the ankle, tarsal joints, and shoulder. Albrecht Wilhelm, in 1959, first described and performed wrist joint denervation for arthritis in a patient with a scaphoid nonunion in Germany. Subsequently, numerous authors have reported their results with Wilhelm’s procedure or variations thereof. Wilhelm and others describe a “complete,” “total,” or “extensive” denervation, sectioning all of the anatomically described articular branches, whereas Weinstein and Berger, in 2002, reported good results after partial wrist denervation in which only the posterior and anterior interosseous articular nerve branches are sectioned. Dellon and associates reported good results with sectioning only of the anterior interosseous nerve (AIN).
Many nerve branches have been described as innervating the wrist joint. The three major nerves of the forearm are the median, ulnar, and radial nerves.
The median nerve main trunk as it passes from the distal forearm to the carpal tunnel does not itself provide any articular branches; rather, it innervates the joint by virtue of its anterior interosseous and palmar cutaneous branches.
The ulnar nerve innervates the ulnar wrist by means of both a small branch from the dorsal sensory ulnar nerve and branches arising from the deep motor branch of the ulnar nerve. It gives off an articular branch in the region of the hook of hamate as well as other branches that pass through the intermetacarpal spaces and emerge dorsally to innervate the carpometacarpal joints of the index to small fingers.
The radial nerve innervates the wrist joint through the posterior interosseous nerve (PIN) and via small branches from the radial sensory nerve.
Other, less well-known nerve branches to the wrist joint include the lateral antebrachial cutaneous nerve (LABCN), which has two tiny branches accompanying the radial artery and venae comitantes, and small articular branches arising from the posterior and medial antebrachial cutaneous nerves.
Classic descriptions of wrist denervation involve multiple small incisions, with elimination of the AIN fibers via cauterization of the distal pronator quadratus. Weinstein and Berger described sectioning only the PIN and AIN via a single longitudinal dorsal incision as a stand-alone treatment for dorsal wrist pain. The AIN itself was identified and removed, as opposed to cauterizing the pronator quadratus.
My current approach to wrist denervation for dorsal and radial wrist pain uses a modification of Berger’s approach to access and transect the PIN and AIN. Through that dorsal incision, branches of the superficial radial and ulnar nerves may also be visualized and removed as needed. A volar longitudinal incision is also used to transect branches of the LABCN running from the radial arterial bundle to the joint, and to ensure that the radial side of the wrist is free of any articular innervating branches from the radial or lateral antebrachial cutaneous nerves. The nerves are eliminated using bipolar cautery, and extreme care is taken to prevent sensory neuromas ( Fig. 61-1 A and B).
INDICATIONS FOR DENERVATION
The ideal candidate is a skeletally mature individual with a chronic, painful radiocarpal wrist condition causing sufficient wrist pain to warrant a surgical procedure after exhausting nonoperative methods. I have not performed denervation as an isolated procedure on skeletally immature individuals, but occasionally have denervated the PIN in adolescents undergoing concomitant excision of a painful dorsal ganglion cyst. After skeletal maturity, there is no upper age limit.
The indications for a stand-alone wrist denervation procedure typically include arthritis due to previous trauma, SLAC (scapholunate advanced collapse) or SNAC (scaphoid nonunion advanced collapse) wrist, and Preiser’s disease. Less common indications include rheumatoid arthritis and Kienböck’s disease. On physical examination, typically, wrist tenderness is present dorsally and radially. Wrist motion is generally reduced, and there may be characteristic dorsal swelling.
In theory, any painful wrist condition not amenable to being fixed and for which salvage procedures are being considered is an indication for wrist denervation. Isolated causes of ulnar-sided wrist pain (e.g., triangular fibrocartilage tears) have not been treated with a stand-alone wrist denervation, but denervation is a possible adjunct to other concomitantly performed procedures. The stand-alone wrist denervation is offered as an alternative to procedures such as partial or complete wrist fusions, proximal row carpectomy, and radial styloidectomy.
Since wrist denervation as a stand-alone procedure is performed for chronic wrist pain, the painful condition has typically been present for at least 6 months to a year before consideration of surgery. Denervation of the PIN is frequently combined with acute procedures as a method of alleviating future wrist pain (e.g., ganglion excisions, wrist fractures). There is no expiration date for performing denervation.
PREOPERATIVE DIAGNOSTIC INJECTIONS
Weinstein and Berger found a poor correlation between the results of preoperative diagnostic PIN/AIN nerve blocks and ultimate postoperative pain or DASH (disabilities of the arm, shoulder, and hand) scores. I perform blocks of the AIN/PIN in the office to see whether any pain relief is afforded. If no pain relief is reported, I usually do not recommend wrist denervation. On the other hand, if pain relief is produced, patients are cautioned that they might not ultimately experience the same degree of pain relief after surgery.
Absolute contraindications to denervation are active infection, mental instability or unreliability, and correctable chronic wrist conditions, such as subacute scaphoid nonunion, or acute conditions that are correctable.
Relative contraindications to denervation are patients with diffuse arthritis, such as severe rheumatoid arthritis; those with unrealistic expectations; those who desire a definitive operation (e.g., arthrodesis) with minimal chance of necessity of revision; and those who show no improvement with preoperative diagnostic injection.
The following are basic requirements for the denervation technique:
Instrumentation: Standard hand instrumentation, loupe magnification, and a bipolar cautery.
Patient position: Supine with arm on hand table.
Tourniquet: Standard nonsterile tourniquet over appropriate padding and protected with a plastic drape from the scrub solution.
Anesthesia: When only denervation of the PIN and AIN with a single dorsal longitudinal incision is performed, local infiltration anesthesia with sedation may be used. For more extensive denervations requiring an additional volar longitudinal incision, regional or general anesthesia is recommended.
Incision Placement and Dissection
Currently, I perform wrist denervation with dorsal and volar longitudinal incisions when denervation is performed as a stand-alone procedure. When performed in conjunction with other dorsal wrist procedures, such as dorsal ganglion excision, or with salvage wrist operations, I usually perform a denervation of the PIN only. I do not attempt to transect the perforating branches of the ulnar nerve via incisions at the base of the metacarpals.
The dorsal incision is longitudinally situated between the radius and ulna from the radiocarpal joint distally to approximately 3 to 5 cm proximal to the ulnar head. The deep forearm fascia is incised, and the interval between the extensor digitorum communis (EDC) and the extensor digiti minimi (EDM) is used ( Figs. 61-2 and 61-3 ).
PIN identification: The EDC is retracted radially and the PIN identified on the floor of the fourth dorsal compartment ( Fig. 61-4 ). We have found the PIN to be on average 0.87 mm in diameter and is equally likely to be radial or ulnar to the posterior interosseous artery. One to two cm of the PIN are resected, and the proximal end is slowly coagulated with bipolar cautery ( Figs. 61-5 and 61-6 ).
AIN identification: The interosseous membrane is incised, exposing the dorsal surface of the pronator quadratus muscle. The AIN is more easily identified at the proximal end of the incision ( Fig. 61-7 ). The AIN is usually found ulnar to the PIN and is radial to the anterior interosseous artery 80% of the time. The average AIN diameter is 1.5 mm at the proximal end of the incision, and it gives off about four motor branches, the largest being the most proximal branch. The first branch comes off about 4 cm proximal to the ulnar head, the last about 2 cm proximal to the ulnar head. The distal sensory branch, which is the branch that is transected, is small—about 0.6 mm. In 40% of cases, there is a 0.4-mm branch to the distal radioulnar joint as well. Therefore, resecting a 1-cm section of the AIN at a level no farther proximal than 2 cm to the ulnar head should denervate its articular branches without denervating the pronator quadratus muscle ( Fig. 61-8 ).